Automatic Tissue Classification for the Human Head from Multispectral MRI
Tolga Taşdizen, David M. Weinstein, James N. Lee · 2004
A robust, automatic algorithm is presented for classifying multispectral MR scans of the human head into nine tissue classes. User initialization is not required. The proposed algorithm is based on three principles: (i) a combination of scalar and multispectral Expectation-Maximization clustering stages are used for fast, unsupervised statistical analysis, (ii) the minimal required amount of prior anatomical knowledge is used to solve statistical ambiguities, and (iii) a top-down modular approach that breaks the problem into smaller logical pieces is adopted.